ZMIME

Ripple-65B Preview vs Hy4 vs SWE-2 Flash

Compare documented capabilities and access. A dash means a comparable figure is unavailable.

Model specifications
SpecificationRipple-65B PreviewRicursive IntelligenceHy4TencentSWE-2 FlashCognition
TaskLanguage generation and coding—Software engineering
Published capabilitiesDense transformer trained on 12 trillion tokens with a focus on long-context software engineering.A next-generation multimodal productivity model with significantly enhanced Agent and complex task execution capabilities.Coding agent model developed from Inkling-Small, with mid-training on scientific and software engineering data.
Model size65B parameters——
AccessPublic weights and hosted inference via Together AI, OpenRouter and Fireworks—Devin Desktop and CLI
SourcesPublisher documentation · Model weights—Publisher documentation
InputsTextText, ImagesText
Context window1,048,576 tokens1,048,576 tokens—
Input price per million tokens—$0.834—
Output price per million tokens—$2.50—
WeightsOpen weightsProprietaryProprietary
ReleasedNov 4, 2026Nov 4, 2026Nov 4, 2026